On 2026-04-28, John Wiegley wrote:
> So, after a few months of working with Claude, C++Ledger is now down to 1
> issue remaining: the desire to offer a PyPI package for Python users.
>
> Now that the dust has settled, please if anyone encounters issues are broken
> behaviors resulting from all of this hectic churn, report them now so that I
> can be sure this version is in good shape before tagging the next release
> candidate.
>
> Thanks and apologies for the turmoil!
>   John
>

Thanks for being patient with my reports, I just checked and I filed 15!

I think a major change is the rounding changes... IIUC ledger used to
always maintain maximum possible precision and then just round at the
end when an amount needed to be displayed. The next release rounds after
every single transaction.

I'm not really sure of the benefit, but the obvious problem is that
people with lots and lots of commodity trades might notice balances
drift after they upgrade. Today, even with the most complex ledger if
you do something like `ledger --cost -X USD bal` the total should be no
more than +/- 1 cent (from the final rounding for display...right?).

I think with the next release, there's no real limit to how far away
from zero it will be. I guess it doesn't break anything, but it feels
like a big change because it's no longer true that "the journal must
always balance to zero"... which used to be the fundamental rule of
ledger...

Tavis.

-- 
 _o)            $ lynx lock.cmpxchg8b.com
 /\\  _o)  _o)  $ finger [email protected]
_\_V _( ) _( )  @taviso

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"Ledger" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/ledger-cli/10tqhgi%24ms4%241%40ciao.gmane.io.

Reply via email to